Self-help organizations for alcohol and drug problems: Toward evidence-based practice and policy

transcript

Journal of Substance Abuse Treatment 26 (2004) 151–158

Regular Article

Self-help organizations for alcohol and drug problems:

Toward evidence-based practice and policy

Keith Humphreys, Ph.D.*, Stephen Wing, M.S.W., Dennis McCarty, Ph.D.,John Chappel, M.D., Lewis Gallant, Ph.D, Beverly Haberle, MHS, A. Thomas Horvath, Ph.D.,

Lee Ann Kaskutas, Dr. P.H., Thomas Kirk, Ph.D., Daniel Kivlahan, Ph.D.,Alexandre Laudet, Ph.D. Barbara S. McCrady, Ph.D., A. Thomas McLellan, Ph.D.,

Jon Morgenstern, Ph.D., Mike Townsend, M.S.S.W., and Roger Weiss, M.D.

Substance Abuse and Mental Health Services Administration/Veterans Health Administration Workgroup on Substance Abuse Self-Help Organizations,

c/o Program Evaluation and Resource Center, Veterans Affairs Health Care System (152-MPD), 795 Willow Road, Menlo Park, CA 94025, USA

Accepted 14 July 2003

Abstract

This expert consensus statement reviews evidence on the effectiveness of drug and alcohol self-help groups and presents potential

implications for clinicians, treatment program managers and policymakers. Because longitudinal studies associate self-help group

involvement with reduced substance use, improved psychosocial functioning, and lessened health care costs, there are humane and practical

reasons to develop self-help group supportive policies. Policies described here that could be implemented by clinicians and program

managers include making greater use of empirically-validated self-help group referral methods in both specialty and non-specialty treatment

settings and developing a menu of locally available self-help group options that are responsive to client’s needs, preferences, and cultural

background. The workgroup also offered possible self-help supportive policy options (e.g., supporting self-help clearinghouses) for state and

federal decision makers. Implementing such policies could strengthen alcohol and drug self-help organizations, and thereby enhance the

Keywords: Self-help groups; Mutual help organizations; Twelve steps; Effectiveness research; Policy

1. Introduction

Self-help organizations are an important resource for

addressing substance abuse, a serious public health problem

that contributes to 500,000 deaths and over $400 billion in

economic costs in the United States each year (Horgan,

Skwara, Strickler, 2001). This white paper summarizes key

research findings on addiction-related self-help groups and

assesses their implications for direct service providers, treat-

ment programs, state agencies and policymakers. This paper

is drawn primarily from the conclusions of a workgroup of

national experts on substance abuse self-help organizations

doi:10.1016/S0740-5472(03)00212-5

* Corresponding author. Tel.: +1-650-617-2746; fax: +1-650-617-

E-mail address: KNH@Stanford.edu (K. Humphreys).

that met November 6–7, 2001, in Washington, D.C. The

information from the workgroup was supplemented by

review of scientific publications, and by the comments of

workgroup participants, observers, self-help group members,

and other stakeholders on earlier drafts of this report.

1.1. Terminology

Addiction and addiction-related refer to all substance-

related problems, including dependence on alcohol, illicit

drugs, or nicotine, as well as being in a close relationship

with a person who has such problems (e.g., a spouse or

parent). Self-help group/organization refers to non-profes-

sional, peer-operated organizations devoted to help-

ing individuals who have addiction-related problems. The

term ‘‘mutual help group’’ is also sometimes used to reflect

the fact that group members give and receive advice,

Table 1

Estimated U.S. membership of selected addiction-related self-help

organizations

Estimated U.S. Membership

Alcoholics Anonymous 1,160,000

Al-Anon Family Groups 200,000

Narcotics Anonymous 185,000

Adult Children of Alcoholics 40,000

Cocaine Anonymous 15,000

Marijuana Anonymous 10,000

Oxford House 9,000

Nicotine Anonymous 7,500

Secular Org. for Sobriety 3,000

K. Humphreys, S. Wing / Journal of Substance Abuse Treatment 26 (2004) 151–158152

encouragement, and support. Self-help groups do not charge

fees and should not be equated with professional treatment

services. Twelve-step organization refers to those self-help

groups that rely on a particular philosophy of recovery that

emphasizes the importance of accepting addiction as a

disease that can be arrested but never eliminated, enhancing

individual maturity and spiritual growth, minimizing self-

centeredness, and providing help to other addicted individ-

uals (e.g., sharing recovery stories in group meetings,

sponsoring new members). Alcoholics Anonymous (AA)

and Narcotics Anonymous (NA) are the best known of the

subset of self-help organizations that rely on the 12 steps.

SMART Recovery 2,000

Women for Sobriety 1,500

Dual Diagnosis Anonymous 700

Note: Data are drawn fromWhite andMadara (2002) andHumphreys (2004).

2. The nature and status of addiction self-help

organizations in the U.S.

Americans participate in a variety of self-help groups for

chronic health problems, including Alzheimer’s disease,

diabetes, cardiovascular disease, obesity, and serious mental

illness. About 18% of American adults have ever attended a

self-help group and about 7% have done so in the previous

12 months (Kessler, Mickelson, & Zhao, 1997). Addiction-

related problems are clearly the most common motivator for

self-help group attendance (Kessler et al., 1997; Room &

Greenfield, 1993). In fact, Americans make more visits to

self-help groups for substance abuse and psychiatric prob-

lems than they do to all mental health professionals com-

bined (Kessler et al., 1997).

Table 1 lists some representative addiction-related self-

help organizations in the U.S. The largest and best known is

AA, a 12-step organization founded in 1935 that inspired

the creation of many of the other organizations listed.

Although the substance and population they address varies,

all the organizations with ‘‘Anonymous’’ or ‘‘Anon’’ in their

name employ a 12-step approach to recovery, as does

Oxford House, a peer-based residential setting, and Double

Trouble in Recovery, a self-help organization for addicted

individuals who also have a serious mental illness. Although

smaller and less well known, the following non 12-step self-

help organizations represent alternatives for substance de-

pendent individuals (Humphreys, 2004):

Secular Organization for Sobriety embraces rationality

and scientific knowledge and does not include any spiritual

content. The organization believes that abstinence can be

achieved through group support and through making sobriety

one’s priority in life.

SMART Recovery views excessive use of alcohol and

other drugs as a maladaptive behavior rather than a disease.

Its goal is to use scientifically informed cognitive-behavioral

techniques to enhance members’ motivation to abstain,

ability to cope with cravings, capacity to identify and modify

irrational thinking, and judgment to balance momentary and

enduring satisfactions.

Women For Sobriety was founded in 1976 to help

women alcoholics recover through a positive, feminist

program that encourages increased self-worth and enhanced

emotional and spiritual growth. It emphasizes the value of

having all-female groups to improve members’ self-esteem

and facilitate their self-discovery.

Another mutual help organization may present an alter-

native for those who abuse alcohol but are not dependent on

it. Moderation Management is a self-help group network of

about 500–1000 people who have decided to reduce or stop

their drinking and make other positive lifestyle changes.

Founded in 1993, it operates under the premise that prob-

lem drinking, unlike chronic alcohol dependence, is a

learned behavioral habit that can be brought under control.

Moderation Management is the only organization men-

tioned in this document that generally attracts individuals

with relatively minor (non-dependent) substance problems

(Humphreys & Klaw, 2001).

In addition to varying in approach, philosophy, and size,

self-help organizations also vary in their governance struc-

ture, organizational traditions (e.g., willingness to accept

outside financial support, encouragement of lifetime mem-

bership) and racial and ethnic diversity. These differences

notwithstanding, none of the above organizations charge

fees, require appointments, or place limits on number of

visits. Members can therefore attend them indefinitely if

they wish. This point is critically important in light of the

emerging conception that addiction is best treated as a

chronic health problem, akin to diabetes and hypertension

in its desired management (McLellan, Lewis, O’Brien, &

Kleber, 2000). Acute care interventions (e.g., hospitaliza-

tion) are important for addressing immediate medical needs,

for stabilization, and for encouraging engagement in con-

tinuing care, but they do not in themselves cure chronic

health problems. Rather, chronic health problems are man-

aged by lower intensity supports over time (Humphreys &

Tucker, 2002). Self-help groups are an important enduring

support for recovery from the chronic health problem of

substance dependence, and complement rather than compete

with acute care interventions.

K. Humphreys, S. Wing / Journal of Substance Abuse Treatment 26 (2004) 151–158 153

A final important point about self-help organizations is

that their growth can be fostered or limited by external

forces. For example, AA experienced a significant increase

in membership in 1941 after The Saturday Evening Post

described it in a highly favorable article. More generally,

non-profit self-help clearinghouses have referred many

potential members to self-help groups and have facilitated

the founding of many groups. Clinician referrals also result

in a significant number of people affiliating with self-help

organizations; similarly, negative clinician attitudes can

discourage participation. And finally, a number of countries,

including Australia, Canada, Germany, Poland, and Japan

have provided funding for the infrastructure of self-help

organizations and have successfully promoted their growth

(Humphreys, 2004).

2.1. Summary of status of U.S. self-help groups

Several conclusions emerge from this description of self-

help organizations that have implications for clinicians,

agencies and policy makers:

� A diverse set of self-help organizations has developed

for all substances of significant public health concern.� Collectively, these self-help organizations are both

appealing and affordable to a broad spectrum of people.� Clinical, agency, and governmental procedures and

policy influence the prevalence, organizational stability,

and availability of addiction-related self-help groups.

3. Research on the effectiveness and cost-effectiveness of

addiction-related self-help organizations

The effectiveness of interventions for substance abuse

must be understood in light of two facts. First, like other

chronic health problems (McLellan et al., 2000), addictive

disorders are difficult to resolve and no intervention produ-

ces complete and permanent abstinence in all cases, or even

in a majority of them. Second, financial resources for

addiction treatment are always constrained, such that any

judgment about whether an intervention is valuable needs to

consider its costs as well as its effectiveness.

The ‘‘effectiveness’’ of self-help organizations can be

conceptualized in a number of ways (e.g., how fast an

organization grows, how it handles change, whether its

members are satisfied with it). However, most clinicians,

agency directors and policy makers are interested primarily

in two specific questions about effectiveness: (1) Does self-

help group participation reduce substance abuse and if so at

what fiscal cost? (2) Do self-help groups benefit members

and society in addition to potentially reducing substance use

per se?

Most outcomes research on addiction mutual help groups

focuses on AA, with NA being the next most commonly

studied organization. Very little outcomes research has been

conducted on non 12-step based self-help groups. An

additional important caveat is that almost every study in

this area has been conducted on adults, leaving the possible

effects of groups on adolescent substance users a much-

understudied question.

Although considered by some to be the most rigorous

scientific test of effectiveness, there are only three random-

ized controlled trials of community-based self-help groups.

All were conducted on AA and all used coerced samples.

The first, conducted in the late 1960s, showed that, com-

pared with individuals assigned to a treatment program or

no treatment, a court order to attend five AA meetings

did not reduce number of arrests for chronic drunkenness

(Ditman, Crawford, Forgy, Moskowitz, & Macandrew,

1967). Unfortunately, this study gathered no information

on alcohol use per se. The other two trials studied a range of

outcomes, and compared AA alone to professional treat-

ments combined with AA attendance (Brandsma, Maultby,

& Welsh, 1980; Walsh et al., 1991). Both suggested worse

clinical outcomes for AA alone, one in terms of more

individuals dropping out and the other in terms of number

of relapses over time. At the same time, individuals assigned

to AA alone in both of these trials improved in absolute

terms from baseline, and had significantly lower health care

costs over time than did those individuals assigned to

treatment plus AA.

Because randomized trials involve only a small and

unrepresentative subset of addicted patients, some research-

ers have conducted quasi-experiments, i.e., assessed out-

comes over time among otherwise similar individuals who

did or did not become involved in an addiction-related

mutual help group. Using this approach, one research team

studied 887 substance dependent patients treated in inpatient

programs that strongly emphasized the importance of

12-step self-help group involvement with those of 887

individuals treated in inpatient programs that had no such

emphasis (Humphreys & Moos, 2001). At treatment intake,

the two groups of patients were comparable on treatment

history, alcohol and drug problems, psychiatric problems,

demographic variables, and motivation. At one-year fol-

lowup, those who were encouraged to join self-help groups

were significantly more likely to be abstaining from drugs

and alcohol. Further, these patients also relied more on self-

help groups and less on further treatment services for

support after discharge, reducing their health care costs by

almost $5000 a year. This study was conducted on male

patients, most of whom were African-American or Hispanic.

Hence, it is worth mentioning that very similar clinical

outcomes and cost-offset findings were found in a separate

study conducted with several hundred alcohol-abusing indi-

viduals, most of whom were Caucasian and about half of

whom were female (Humphreys & Moos, 1996).

A third type of study examines the correlation between

self-help group involvement and substance use, with or

without a comparison group of non-participants, and some-

times without a longitudinal design for tracking change over

time. Almost all such studies find that AA attendance is

associated with better alcohol-related outcomes (e.g., re-

duced consumption, fewer alcohol dependence symptoms),

and that NA and Cocaine Anonymous attendance is asso-

ciated with better drug-related outcomes (for a review, see

Humphreys, 2004). These same studies also show that

members who engage in other group activities in addition

to attending meetings—reading program literature, sponsor-

ing new members, applying the 12 steps to daily life—are

more likely to abstain from substances than are individuals

who do not engage in these activities. Yet such correlational

studies do not prove that the self-help group caused the

positive outcome from a purely scientific standpoint; they

show only that there was a positive outcome.

Finally, although treatment is not a self-help group, some

studies of self-help influenced treatments provide relevant

evidence. The best known of these studies is Project

MATCH, which randomly assigned alcohol dependent

patients to one of three treatments delivered over a three-

month period (Babor & Del Boca, 2002; Project MATCH

Research Group, 1997). The treatments were manual-guided

12-step facilitation therapy, cognitive-behavior therapy, and

motivational enhancement therapy. One-year post-treatment

outcomes were largely similar for patients in all three

conditions in terms of increased days of abstinence, as well

as reduced average number of drinks consumed per day.

Individuals treated in 12-step facilitation therapy attended

significantly more 12-step self-help group meetings and

were more likely to have maintained continuous abstinence.

Patients receiving 12-step facilitation continued to have

higher rates of continuous abstinence three years after

treatment, and, when compared with patients receiving

cognitive behavioral therapy, had a greater percentage of

abstinent days. Finally, regardless of assigned treatment

condition, more 12-step self-help group attendance was

associated with better outcomes.

Encouraging results were also found in a major clinical

trial addressing cocaine dependent patients. The Collabora-

tive Cocaine Treatment Study found that patients randomly

assigned to group and individual counseling sessions in

which they were strongly encouraged to attend self-help

groups showed more consistent attendance and more con-

secutive months of cocaine abstinence during followup

compared with the other three treatments, which included

only professionally administered therapies (Weiss et al.,

1996, 2000).

Three other studies of self-help influenced treatment

warrant mention. A study of drug dependent patients found

that those randomly assigned to an aftercare program

incorporating a self-help style group and network of sup-

portive former patients were about 40% less likely to relapse

over the next six months compared with those receiving

usual aftercare (McAuliffe, 1990). A second study found

that alcohol dependent patients randomly assigned to an

experimental treatment program that emphasized peer re-

sponsibility and mutual help had higher treatment engage-

ment and dramatically lower health care costs at one-year

followup (Galanter, 1984). A third study randomly assigned

adult substance dependent patients who had been raised by

substance dependent parents to attend either 12-step self-

help groups for Adult Children of Alcoholics or didactic

education classes about substance abuse (Kingree &

Thompson, 2000). Those patients who were assigned to

self-help groups were significantly less likely to use drugs

and alcohol after treatment discharge.

The studies described above suggest that self-help

group involvement reduces substance use and also lowers

health care costs. These and a number of other research

projects have also documented other benefits of self-help

group participation, including enhanced self-efficacy, im-

proved social support, reduced depression and anxiety, and

more effective coping with stress (see Humphreys, 2003,

for a review). The benefits of addiction self-help groups

thus seem to extend beyond reductions in substance use

per se.

The research cited above focuses on AA and NA. Many

of the findings may generalize to other mutual help organ-

izations. Research has not been undertaken to date to

investigate this hypothesis, however.

3.1. Synthesis of effectiveness research results

A significant body of research has documented an

association between 12-step self-help group participation

and positive outcomes and has suggested mechanisms by

which these positive outcomes are generated. In addition,

millions of Americans have ‘‘voted with their feet’’ by

participating in addiction-related self-help groups, some-

times in the face of ambivalence by clinicians. Many

improvements remain to be made in self-help group re-

search, but at present the following represent reasonable

conclusions based on the existing research:

� Longitudinal studies associate Alcoholics Anonymous

and Narcotics Anonymous participation with greater

likelihood of abstinence, improved social functioning,

and greater self-efficacy. Participation seems more

helpful when members engage in other group activities

in addition to attending meetings.� Twelve-step self-help groups significantly reduce

health care utilization and costs, removing a significant

burden from the health care system.� Self-help groups are best viewed as a form of

continuing care rather than as a substitute for acute

treatment services (e.g., detoxification, hospital-based

treatment, etc.).� Randomized trials with coerced populations suggest that

AA combined with professional treatment is superior to

AA alone.� Non 12-step self-help groups have not been subjected to

longitudinal outcome evaluation but it is reasonable to

suspect they also benefit members.

4. Potential implications for clinicians, treatment

programs, and policymakers

The foregoing indicates that addiction-related mutual

help organizations likely contribute significantly to public

health and also lower health care costs. Clinicians, treatment

providers, and policymakers may therefore wish to develop

and implement practices and policies that increase the

likelihood that addicted individuals will seek out mutual

help organizations and that these organizations will become

more prevalent and accessible to a broad array of people.

The remainder of this paper presents potential courses of

action, focusing first on clinicians and treatment programs,

and then turning to policymakers.

4.1. Clinicians and treatment program directors

Many people believe that all American substance abuse

treatment programs are based on the 12 steps and that all

clinicians in them are already promoting self-help groups, so

there is really nothing that could be done to make treatment

programs better at facilitating self-help group involvement.

Research does not support these assumptions. Even practi-

tioners who describe themselves as ‘‘12-step oriented’’ con-

sider only a subset of 12-step processes important for clients.

Further, few professionals report operating from a pure 12-

step approach, preferring instead an eclectic mix of

approaches, e.g., 12-step, cognitive-behavioral, motivational

interviewing, psychodynamic, family systems, etc. These

findings have been confirmed in videotape studies of psy-

chotherapy sessions, which show counselors emphasizing

some aspects of the 12 steps, such as AA affiliation, and not

emphasizing others, such as spirituality (Morgenstern &

McCrady, 1993). When counselors do attempt to support

12-step self-help group involvement, they rarely use empir-

ically-supported methods. Finally, many clinicians are not

even aware of alternatives to 12-step self-help groups. There

is thus a great deal of improvement to be made in these areas.

Research has clearly demonstrated that when clinicians

use empirically-validated techniques to support mutual help

group involvement, it is far more likely to occur (Nowinski,

Baker, & Carroll, 1995; Sisson & Mallams, 1981; Weiss

et al., 1996, 2000). Educating clinicians about such techni-

ques may be helpful in some cases, but in general, merely

providing empirically-supported treatment guidelines to

clinicians rarely changes their practice patterns significantly.

Changing the behavior of clinicians is a significant chal-

lenge upon which addiction researchers are only beginning

to focus significant attention.

In addition to changing clinical behavior, efforts to

promote self-help group affiliation must also consider clini-

cian beliefs that influence patients’ transition from treatment

to self-help groups. Some treatment providers see self-help

groups negatively, positing that they may foster unhealthy

dependence or detract from personal autonomy. Other pro-

viders think that AA is the only self-help organization that

exists or is the only intervention of any value. Other

misconceptions include the belief that all self-help organ-

izations have a spiritual component, or that spirituality must

be central for every member of AA, NA, and other 12-step

groups. In reality, there are many pathways to recovery

involving a variety of self-help groups and treatments

(Fletcher, 2001). Hence, provider education must address

both attitudes and behaviors in order to create a successful

interface of clinicians with a broad self-help group network.

Any professional education strategy along these lines

must recognize two important points. First, most investiga-

tions have focused on specialty substance abuse treatment

providers and researchers thus know little about how non-

specialty providers (e.g., emergency room physicians) refer

addicted patients to self-help groups, or for that matter if they

do so at all. Second, any effort to change clinical behavior,

regardless of the treatment specialty concerned, must be

sensitive to the diversity of patients. Some substance abuse

self-help organizations, for example, SMART Recovery,

Women for Sobriety, and Moderation Management, have

an almost entirely Caucasian, middle class membership. The

membership of AA and NA includes a higher proportion of

people of color, but individual chapters of these organizations

may not necessarily be diverse. Clinicians should be sensitive

to potential patient discomfort among patients going to a self-

help group with few or no people of their racial or ethnic

background. Relatedly, gay and lesbian patients may prefer

specialty meetings (such as AA offers), and clinicians should

be aware of this and also of where such meetings are held. In

summary, because of cultural differences (e.g., in spiritual

beliefs, expectations about self-disclosure) and other diver-

sity issues, all self-help organizations may not be equally

appealing or helpful to all patients. These and other diversity

issues need to be thoughtfully addressed.

4.2. Potential strategies

The following strategies could be employed by individ-

ual clinicians, clinical supervisors, and program directors:

� Clinicians should use empirically-validated methods

(e.g., 12-step facilitation counseling, motivational

enhancement techniques) when seeking to foster self-

help group engagement.� Given the variety of pathways to recovery, clinicians

should have a menu of treatment and self-help group

options available for use when selecting care alter-

natives in consultation with the client and other

stakeholders.� Efforts to train clinicians about facilitating self-help

group involvement should include incentives for

changing clinical practice and should be sensitive to

cultural diversity.� Effective referrals to addiction self-help groups

should occur in both non-specialty and specialty

health care programs.

� Clinicians should recognize and communicate to

patients that many individuals recover through AA,

but others recover through self-help groups other than

AA, or, without attending any self-help group at all.� Even treatment programs that label and represent

themselves as ‘‘12-step oriented’’ should evaluate

whether their current program practices actively

support involvement in 12-step self-help groups.

5. Potential implications for policymakers

A number of countries outside of the U.S. have imple-

mented policies designed to foster the growth of mutual help

organizations (Humphreys, 2004). In the U.S., some treat-

ment systems, states and federal agencies have also attemp-

ted such initiatives. As with all policies, efforts in this area

will face challenges at both the state and national level

related to funding, coordination, and implementation. In

addition, there is an concern peculiar to this area. Any

efforts to support self-help organizations must recognize that

by tradition, 12-step organizations do not accept direct

outside financial support. Even for self-help organizations

that do, it is important not to bureaucratize or co-opt what is

essentially a grassroots movement. These challenges are

worth meeting because of the potential for self-help groups

to be a cost-effective intervention for addiction.

Like self-help organizations themselves, the self-help

supportive infrastructure varies in strength and organization

from place to place. Non-profit organizations known as ‘‘self-

help clearinghouses’’ exist in some areas, and provide infor-

mation about, referrals to, and technical support for mutual

help organizations for addictions and other health problems.

Help lines and human service agencies in some parts of the

country also provide information on self-help groups.

‘‘Recovering’’ counselors and alumni groups at addiction

treatment centers are an additional important component of

self-help group-supportive infrastructure. In the wake of

managed care and changes in credentialing rules, such

potential sources of support for self-help groups may be

weakening in strength, thereby requiring other concerned

health care professionals to become involved in the process.

Whether individuals not in recovery have the knowledge

and skills to facilitate connections between addicted patients

and self-help groups is unknown.

5.1. Potential strategies

Given the above context, it may be desirable to imple-

ment policies that could strengthen the infrastructure sup-

porting mutual help organizations. The following policy

efforts have been implemented in some nations or states,

and might be replicated in other contexts.

� Invest resources in self-help clearinghouses. These

organizations can support a broad variety of alcohol

and drug-related self-help groups without violating the

traditions of those that do not accept funding.� Make public facilities and institutions self-help group

friendly. This includes not only allowing groups to use

space for meetings, but also inviting them to hold

groups in settings where they may not have a historical

presence, for example some clinics, hospitals, religious

organizations, and community centers.� Disseminate information on self-help groups. Govern-

ment agencies and interested non-governmental organ-

izations could post lists of self-help organizations on

their web sites and/or provide links to web sites operated

by self-help organizations that provide such informa-

tion. Such dissemination efforts could also provide

information on evidence-based practices related to self-

help groups as a recovery resource.� Adopt the principle of ‘‘informational parity.’’ Dissem-

ination efforts of all forms should include information

on the full range of mutual help group alternatives. As

long as mutual help groups are voluntary in nature,

respect the civil rights of participants, address substance

abuse, are not professional treatments mislabeled as

self-help groups, and have some evidence of effective-

ness, they should be included on listings of drug and

alcohol self-help groups.� Create and support innovative services that promote

self-help group involvement. Examples include the

recovery coaches funded through the Arizona Medic-

aid program and a program in Philadelphia that

provides funds to an organization that accepts

responsibility for transitioning the individual into

self-help groups. Similarly, SAMHSA’s Center for

Substance Abuse Treatment (CSAT) launched the

Recovery Community Services Program to provide

funding to groups who are developing innovative peer-

to-peer services. Examples of services under this new

program include recovery coaching and mentoring,

peer case management, peer education in life skills

(e.g., parenting, communication) and health topics,

assistance and referral with housing, employment,

education, and related activities.� Credential and train healthcare professionals in linking

patients to self-help groups. Because staff with strong

connections to local self-help groups may not be

present in all settings, all health care professionals

should have some knowledge about how to refer

patients effectively to groups.� Foster self-help organizations for underserved popula-

tions. New York State’s Mental Health Empowerment

Project successfully assisted consumers of mental

health services to organize self-help groups for

dually-diagnosed people. Similar programs, that pro-

vide support without professionalizing or bureaucratiz-

ing self-help groups, might be tried with other

underserved groups, such as adolescents and residents

of rural areas.

� Expand opportunities for self-help organizations in

criminal justice settings. Self-help groups can be made

available to offenders in conjunction with treatment in

correctional facilities and in the community. For

example, invitations might be given to groups to hold

meetings in juvenile detention facilities, jails, prisons,

probation services, and parole departments. Given the

coercive nature of treatment in criminal justice

settings, program directors and clinicians should avoid

forcing clients to participate in self-help groups when

it is not appropriate, and should offer alternatives to

such clients.� Discourage the use of self-help groups as a replace-

ment for treatment. Research shows that many clients

require the support both of treatment programs and of

self-help groups. Using the success of self-help groups

as a pretext for delaying or reducing support for

treatment services is therefore inappropriate. Addic-

tion self-help organizations typically see themselves as

an ally, rather than as a competitor to professional

treatment programs; other stakeholders in this area

should adopt the same perspective.� Expand research on drug and alcohol self-help groups.

Evaluation research on both 12-step and non 12-step

self-help groups should be expanded. So too should

research on the mechanisms through which self-help

groups effect change, and on policy interventions that

might promote technology transfer and self-help group

involvement. Establishing a National Center for Re-

search and Technology Transfer on Self-Help Groups

and Addiction could provide an important focus for

such activities.� Expand residential self-help options. Oxford House is a

national program with over 850 peer-managed houses.

Connecticut and California also have successful resi-

dential models of peer-managed services for addicted

individuals. Fostering the development of more self-

help based housing could be a cost-effective strategy for

providing recovery-supportive environments for sub-

stance dependent individuals, including those who

are homeless.� Support opportunities for family members of addicted

people to be involved in mutual help organizations.

One of the discoveries of CSAT’s recovering commun-

ities program was that families do not always feel a part

of the recovering person’s involvement in a self-help

organization. Accordingly, all of the above efforts

should include a focus on family members and family-

focused mutual help organizations.

6. Conclusions

Addiction self-help organizations are a major resource

for addicted individuals, as well as for those who treat

addicted people, work with them, and care about them.

Research to date suggests that self-help groups can be

beneficial, but also cautions that we have much more to

learn about how they work and how they can be supported

through clinical, agency, and policy actions. The strategies

presented herein are therefore a set of initial steps and are

neither the final word nor a panacea. Yet they do hold

significant promise of strengthening addiction self-help

groups and thereby helping more individuals recover from

drug and alcohol problems.

Acknowledgments

Preparation of this document was supported by the

Substance Abuse and Mental Health Services Administra-

tion (SAMHSA), U.S. Department of Health and Human

Services and the Mental Health Strategic Healthcare Group

and Health Services Research and Development Service,

U. S. Department of Veterans Affairs (VA). Conclusions in

this document do not necessarily represent official views of

SAMHSA, the VA, or the organizations with which any

author or commentator on this paper is affiliated. We are

grateful to the following individuals for commenting on

workgroup deliberations and/or earlier drafts of this docu-

ment: Sonya A. Baker, T. Robert Burke, Herman Die-

senhaus, Dona M. Dmitrovic, R. John Gregrich, Tom Hill,

Mike Hilton, George Kosniak, John Mahoney, Kate

Malliarakis, J. Paul Molloy, Harold Perl, Rick Sampson,

and Richard Suchinsky.

References

Babor, T., & Del Boca, F. (Eds.) (2002). Treatment matching in alcoholism.

Cambridge, UK: Cambridge University Press.

Brandsma, J. M., Maultby, M. C., & Welsh, R. J. (1980). Outpatient treat-

ment of alcoholism: A review and comparative study. Baltimore, MD:

University Park Press.

Ditman, K. S., Crawford, G. G., Forgy, E. W., Moskowitz, H., & Macan-

drew, C. (1967). A controlled experiment on the use of court probation

for drunk arrests. American Journal of Psychiatry, 124, 64–67.

Fletcher, A. (2001). Sober for good: New solutions for drinking problems.

Boston: Houghton Mifflin.

Galanter, M. (1984). Self-help large group therapy for alcoholism: A

controlled study. Alcoholism: Clinical and Experimental Research, 8,

16–23.

Horgan, C., Skwara, K. C., & Strickler, G. (2001). Substance abuse: The

nation’s number one health problem. Princeton, NJ: The Robert Wood

Johnson Foundation.

Humphreys, K. (2004). Circles of Recovery: Self-help organizations for

addictions. Cambridge, UK: Cambridge University Press.

Humphreys, K., & Klaw, E. (2001). Can targeting non-dependent problem

drinkers and providing internet-based services expand access to assis-

tance for alcohol problems? A study of the Moderation Management

self-help/mutual aid organization. Journal of Studies on Alcohol, 62,

528–532.

Humphreys, K., & Moos, R. H. (2001). Can encouraging substance abuse

inpatients to participate in self-help groups reduce the demand for out-

patient aftercare? A quasi-experimental study. Alcoholism: Clinical and

Experimental Research, 25, 711–716.

Humphreys, K., & Moos, R. (1996). Reduced substance abuse-related

health care costs among voluntary participants in Alcoholics Anony-

mous. Psychiatric Services, 47, 709–713.

Humphreys, K., & Tucker, J. (2002). Towards more responsive and effec-

tive intervention systems for alcohol-related problems. Addiction, 97,

126–132.

Kessler, R. C., Frank, R. G., Edlund, S. J., Katz, E., Lin, E., & Leaf, P.

(1997). Differences in the use of psychiatric outpatient services between

the United States and Ontario. New England Journal of Medicine, 326,

551–557.

Kessler, R. C., Mickelson, K. D., & Zhao, S. (1997). Patterns and correlates

of self-help group membership in the United States. Social Policy, 27,

27–46.

Kingree, J., & Thompson, M. (2000). Mutual help groups, perceived status

benefits, and well-being: A test with adult children of alcoholics with

personal substance abuse problems. American Journal of Community

Psychology, 28, 325–342.

McAuliffe, W. E. (1990). A randomized clinical trial of recovery training

and self-help for opiod addicts in New England and Hong Kong. Jour-

nal of Psychoactive Drugs, 22, 197–209.

McLellan, A. T., Lewis, D. C., O’Brien, C. P., & Kleber, H. D. (2000).

Drug dependence, a chronic medical illness: Implications for treatment,

insurance, and outcomes evaluation. Journal of the American Medical

Association, 284, 1689–1695.

Morgenstern, J., & McCrady, B. S. (1993). Cognitive processes and change

in disease-model treatment. In: B. S. McCrady, & W. R. Miller (Eds.),

Research on Alcoholics Anonymous: Opportunities and alternatives ( pp.

153–166). New Brunswick, NJ: Rutgers Center of Alcohol Studies.

Nowinski, J., Baker, S., & Carroll, K. (1995). Twelve-step facilitation ther-

apy manual: A clinical research guide for therapists treating individuals

with alcohol abuse and dependence. Rockville, MD: NIAAA (NIH Pub-

licationNo.94–3722).

Project MATCH Research Group. (1997). Matching alcoholism treatments

to client heterogeneity: Project MATCH posttreatment drinking out-

comes. Journal of Studies on Alcohol, 58, 7–29.

Room, R., & Greenfield, T. (1993). Alcoholics Anonymous, other 12-step

movements and psychotherapy in the US population, 1990. Addiction,

88, 555–562.

Sisson, R. W., & Mallams, J. H. (1981). The use of systematic encourage-

ment and community access procedures to increase attendance at Alco-

holics Anonymous and Al-Anon meetings. American Journal of Drug

and Alcohol Abuse, 8, 371–376.

Walsh, D., Hingson, R., Merrigan, D., Levenson, S., Cupples, L., Heeren,

T., Coffman, G., Becker, C., Barker, T., Hamilton, S., McGuire, T., &

Kelly, C. (1991). A randomized trial of treatment options for alcohol-

abusing workers. New England Journal of Medicine, 325, 775–782.

Weiss, R. D., Griffin, M. L., Gallop, R., Onken, L. S., Gastfriend, D. R.,

Daley, D. C., Crits-Cristoph, P., Bishop, S., & Barber, J. P. (2000). Self-

help group attendance and participation among cocaine dependent pa-

tients. Drug and Alcohol Dependence, 60, 169–177.

Weiss, R. D., Griffin, M. L., Najavits, L. M., Hufford, C., Kogan, J.,

Thompson, H. J., Albeck, J. H., Bishop, S., Daley, D. C., Mercer, D.,

& Siqueland, L. (1996). Self-help activities in cocaine dependent pa-

tients entering treatment: Results from the NIDA collaborative cocaine

treatment study. Drug and Alcohol Dependence, 43, 79–86.

White, B. J., & Madara, E. J. (2002). The self-help sourcebook: Your guide

to community and online support groups (7th ed.). Denville, NJ: Amer-

ican Self-Help Clearinghouse.

Self-help organizations for alcohol and drug problems: Toward evidence-based practice and policy

Documents